Superpixel Sizes using Topology Preserved Regular Superpixel Algorithm and their Impacts on Interactive Segmentation

Abstract

Interactive Image Segmentation is a type of semiautomated segmentation that uses user input to extract the object of interest. It is possible to speed up and improve the end result of segmentation by using pre-processing steps. The use of superpixels is an example of a pre-processing step. A superpixel is a collection of pixels with similar properties such as texture and colour. Previous research was conducted to assess the impact of the number of superpixels (based on SEEDS superpixel aglorithms) required to achieve the best segmentation results. The study, however, only examined one type of input (strokes) and a small number of images. As a result, the goal of this study is to extend previous work by performing interactive segmentation with input strokes and a combination of bounding box and strokes on images from Grabcut image data sets generated by Topology preserved regular superpixel (TPRS). Based on our findings, an image with 1000 to 2500 superpixels and a combination of bounding box and strokes will help the interactive segmentation algorithm produce a good segmentation result. Finally, the size of the superpixels would influence the final segmentation results as well as the input type.

How Does Image Complexity Affect the Accuracy of an Interactive Image Segmentation Algorithm?

“…Finally, it finds the local optimal path between vertical and horizontal seed pairs. In the previous study [7], the MSRM was able to produce an ideal segmentation result with a bounding box with three strokes by employing superpixel sizes of 1000 and 2500. As a result, this configuration will be employed in this study.…”

“…Bounding box [21][22][23]: The user must position the bounding box around the object of interest within the image. In the previous study [7], it was determined that achieving an optimal result in interactive image segmentation can be accomplished by using a bounding box with three strokes. Building upon this insight, our study extends the approach by employing a bounding box as the background and strokes as the foreground in the segmentation process.…”

“…Existing research has examined various factors such as the number of strokes, superpixel sizes [6,7], and different superpixel algorithms [8] on complex images. However, it is worth noting that previous studies often determined the complexity of images using subjective methods.…”